The co-evolution of taxi drivers and their in-car navigation systems

Fabien Girardin, Universitat Pompeu Fabra2008 AAG annual meeting, April 15, 2008

Motivation

• 17% U.S. adults use a GPS location device/service. 33% of them use it in their vehicle (Harris Interactive, 2007)

• Relative market success with issues of poor intergration of the technology into driving practices

• Chance to observer how users adapt to the technology and adapt the technology to their needs (co-evolution - O’Day et al, 1996)

• Observe how positioning and navigation technologies integrate into/change already existing practices and how practices influence the use of the technology

• Inform the design

Aim

• Similar approach as other research on understanding mobility practices with technology devices (Activities not as isolated events, but situated withing a context)

• Socio-technical lenses to analyze the complexities of changes created by the introduction of new technologies

• Acknowledge change and adaptibility as inherent part of location-aware system use

• Focus on drivers for whom mobility is a labor

Previous works

• We know that there are issues in realability, wayfinding, the support of mobile workers

• Interaction with the in-car device result in distraction from driving task (often in the news). Studies at an experimental level (see Burnett’s work)

• However, device-centered perspective falls short of full appreciation of the envolvment people and their use of novel mobility enhancing technologies.

• It neglects the role of the user in co-constructing technology (technology studied in context)

• Lack of empricial evidences. Particularly usefull for the design of future location-aware systems

Focus

Analysis of use, adoption, and appropriation

Not focused on how the system work but how the systems are used

Method

• Interpretative and not statistical

• Examine situated activities

• Ethnography, interviews supported by in-car obsevations

• Users accounts

• Small sample (12 taxi drivers) but, but rich data (extended field notes of accounts and observations). Coded as themes emerges

• 6 months to 25 years of experience

Locus

Barcelona taxi drivers: massive population of early adopters, strong practice of relying on mobile technologies and maps to support their work

Findings

Eco-system of artifacts

Sources of information

• satnav (discussed later)

• “Guia”: usefull in areas with points of reference

• Mobile phone: primarly a social link, used when “really lost”

• Newspaper: keep general knowledge of the activity in the city

• Personal notes: list of “unofficial” POI not present in satnav or “Guia”

• The customers is a prime resource for information

Acquisition of tranquility

• Acquisition of satnav is not about performance, safety, or efficiency

• Biggest moment of uncertainty is to drive to unknown neighboroods (villages, remote business areas)

• Satnav is a life saver for getting to destination

• Relieves the car driver to closely observe the environment, to look out for road signs or landmarks

• Relaxes the customer “they know they cannot be cheated with it”

• HOWEVER, the less experience drivers are confronted to system imperfections

Knowledge of the city

• Different appropriation depending on the knowledge of the space

• Inexperience drivers demand of accurate and updated information quickly reached the limits of the systems (unable to judge the accuracy or improvise)

• Their dependance made them rely on the “guia”

• Satnav as a learning tool (passive mode)

• Tendency to rely on it less and less with the experience (for navigation). Passive mode used to keep track of radars or speed

• Experienced drivers stopped using the “guia”

Breack the myth of deskilling

• Assumption that satnav makes the practice accessible to cheap labor

• Satnav supports a reduced understanding of the city

• This study shows that knowledge is acquired by multiple sources, technology do not replace it

The different granularity of information

• To go: indication of an area, a neighbood. Coarse-grained (no indications to avoid misleading recommandation)

• To arrive: specific builing number. Fine-grained (with indications). Satnav engaged at a stop, slow traffic.

• To return: Need of landmark. Very cautious in the path taken. Push “barcelona” or “plaza espana” to get back to familiar places. Both fine-grained/coarse-grained

Implication for design

• Big issue of uncertainty and its represenation. Reveal the ambiguity of positioning and incomplete map data (important for inexperienced drivers)

• Distinctive usage in a journey. Tailor the navigation to the context (go, arrive and return)

Conclusion

• First chance to observer how users adapt to the technology and adapt the technology to their needs in the context of location-aware systems.

• Limitations in the generalization: spanish drivers (cultural bias), males (gender bias), workers (activity bias)

• Wayfinding, tendency to be used less over time

• Not necessarly deskilling of navigation and orientation

• Maybe deskilling in social. Transfer of the trust from social interacting to machine-supported. May lose interaction with the client, important in the learning process. Or new kind of interaction